PLANT GENETICS • ORIGINAL PAPER

Development of gene-based molecular markers tagging low alkaloidpauper locus in white lupin (Lupinus albus L.)

Sandra Rychel1 & Michał Książkiewicz1

Received: 2 April 2019 /Revised: 2 July 2019 /Accepted: 18 July 2019# The Author(s) 2019

AbstractWhite lupin (Lupinus albus L.) is a legume grain crop cultivated since ancient Greece and Egypt. Modern white lupin cultivarsare appreciated as a source of protein with positive nutraceutical impact. However, white lupins produce anti-nutritional com-pounds, quinolizidine alkaloids, which provide bitter taste and have a negative influence on human health. During domesticationof this species, several recessive alleles at unlinked loci controlling low alkaloid content were selected. One of these loci, pauper,was exploited worldwide providing numerous low-alkaloid cultivars. However, molecular tracking of pauper has been hampereddue to the lack of diagnostic markers. In the present study, the synteny-based approach was harnessed to target pauper locus.Single-nucleotide polymorphisms flanking pauper locus on white lupin linkage map as well as candidate gene sequenceselucidated from the narrow-leafed lupin (L. angustifolius L.) chromosome segment syntenic to the pauper linkage group regionwere transformed to PCR-based molecular markers. These markers were analyzed both in the mapping population and worldgermplasm collection. From fourteen markers screened, eleven were localized at a distance below 1.5 cM from this locus,including five co-segregating with pauper. The linkage of these markers was confirmed by high LOD values (up to 58.4).Validation performed in the set of 127 bitter and 23 sweet accessions evidenced high applicability of one marker,LAGI01_35805_F1_R1, for pauper locus selection, highlighted by the low ratio of false-positive scores (2.5%).LAGI01_35805 represents a homolog of L. angustifolius acyltransferase-like (LaAT) genewhichmight hypothetically participatein the alkaloid biosynthesis process in lupins.

Keywords White lupin . Alkaloids .Molecular markers . Linkagemapping

Introduction

White lupin (Lupinus albusL.) is a cool season grain legumecrop with a relatively long history of cultivation. Primarydomestication of L. albus has occurred in ancient Greeceand Egypt to produce grain for human and animal consump-tion as well as in ancient Rome as green manure (Gladstones1970). Lupins were found to be very beneficial in crop

rotations because they increase soil fertility through symbi-otic nitrogen fixation and efficient mobilization of soil phos-phorus (Lambers et al. 2013). Modern lupin cultivars areappreciated also as a valuable source of protein (38–42% inseeds) (Papineau and Huyghe 2004) with positive nutraceu-tical impact on hypercholesterolemia, hypertension, and hy-perglycemia (Arnoldi and Greco 2011). Moreover, white lu-pin crops have moderate seed content of oil (10–13%) withdesirable ratios of omega-6 to omega-3 acids for consump-tion purposes (Boschin et al. 2007). These advantages makethis species valuable for human food and animal feed (Linet al. 2009). However, white lupin seeds contain some con-tent of anti-nutritional compounds, including quinolizidinealkaloids (up to 12% in wild populations) and oligosaccha-rides (up to 10%) (Kroc et al. 2017; Mohamed and Rayas-Duarte 1995). Average total alkaloid content in white lupinbreeding lines and cultivars is about 1.3% of the seed dryweight; however, in some sweet accessions, these valuesare below 0.02% (Kroc et al. 2017).

Communicated by: Barbara Naganowska

Electronic supplementary material The online version of this article(https://doi.org/10.1007/s13353-019-00508-9) contains supplementarymaterial, which is available to authorized users.

* Michał Książkiewiczmksi@igr.poznan.pl

1 Institute of Plant Genetics, Polish Academy of Sciences,Strzeszyńska 34, 60-479 Poznań, Poland

Journal of Applied Geneticshttps://doi.org/10.1007/s13353-019-00508-9

(2019) 60:269–281

/Published online: 13 2019August

Alkaloids are considered as the major unfavorable compo-nents in white lupin due to their bitter taste and negative in-fluence on human health, causing in the worst-case scenarioacute anticholinergic toxicity (Daverio et al. 2014). Therefore,during the domestication process, numerous efforts aiming atthe reduction of alkaloid levels were strongly emphasized. Asmany as nine hypothetical loci controlling low alkaloid con-tent were initially identified throughout white lupin breeding,including pauper, primus, tercius, exiguus, nutricius, mitis,suavis, reductus, and minutus (Hackbarth 1957; Hackbarth1961; Porsche 1964; Šatović 1993; Troll 1958). However,primus and terciuswere identified as the synonyms of pauper,whereas suavis andminutuswere not studied as extensively asthe other loci and their independence cannot be authenticated(Harrison and Williams 1982). Three low-alkaloid allelesoriginating from different loci were introduced into white lu-pin cultivars in the early years of modern domestication,namely exiguus (cv. Neuland, 1937), pauper (cv. Kraftquell,cv. Ultra, and cv. Przebędowski Wczesny, 1949–1950), andnutricius (cv. Nahrquell, 1949); nevertheless, only a pauperwas exploited worldwide (Harrison and Williams 1982;Šatović 1993).

To facilitate molecular studies on white lupin domestica-tion genes, a recombinant inbred line (RIL) mapping popula-tion was developed, descending from parental cross betweenKiev Mutant (Ukrainian cv, sweet, early flowering, anthrac-nose susceptible) and P27174 (Ethiopian landrace, bitter, lateflowering, anthracnose resistant) (Phan et al. 2007). A linkagemap carrying 220 amplified fragment length polymorphism(AFLP) and 105 PCR-amplified gene–based markers was de-veloped for this RIL population, but the distance between thepauper locus and flanking markers was revealed to be higherthan 20 cM in both directions (Phan et al. 2007). This mapwaslater updated with the set of 136 diversity array technology(DArT) markers but no improvement around the pauper locuswas achieved as the flanking markers were split into two sep-arate linkage groups (Vipin et al. 2013). With the aid of themicrosatellite-anchored fragment length polymorphism(MFLP) technique, the PauperM1 marker was developed,which was confirmed in the Kiev Mutant × P27174 RIL pop-ulation to be linked to the pauper locus at the genetic distanceof 1.4 cM (Lin et al. 2009). However, this marker requiredsequencing gel and radioisotope primer labelling for the cor-rect determination of alleles. Moreover, the applicability ofthis marker was restricted to ~ 95% of bitter lines and 91%of sweet non-pauper lines (Lin et al. 2009). Recently, a high-density consensus linkage map of white lupin genome wasconstructed, which integrated 453 published markers with3597 newly developed sequence-based markers and constitut-ed a single linkage group per every chromosome(Książkiewicz et al. 2017). This map yielded several newmarkers co-segregating or localized closer to the pauper locusthan the PauperM1.Moreover, this new reference linkage map

anchored recently published transcriptome assembly(O’Rourke et al. 2013) to particular markers, and aligned thesemarkers to syntenic blocks of the narrow-leafed lupin(L. angustifolius L.) genome sequence (Hane et al. 2017),providing novel opportunities for tracking white lupin domes-tication genes by comparative mapping approach.

In the present study, these resourceswere harnessed to analyzethe genome region carrying low-alkaloid pauper locus. PCR-based markers were developed and implemented for the screen-ing of white lupin germplasm collection carrying a diversifiedsubset of sweet and bitter lines. The applicability of newly de-veloped markers for pauper allele selection has been evaluated.

Materials and methods

Plant material

Genetic mapping was performed using the reference L. albusKiev Mutant × P27174 recombinant inbred line (RIL) popu-lation (F8, n = 195), delivered by the Department ofAgriculture and Food, Western Australia. This populationwas derived from a cross between a bitter, late flowering,and anthracnose-resistant Ethiopian landrace (P27174) and asweet, early flowering, and anthracnose susceptible Ukrainiancultivar (Kiev Mutant) (Książkiewicz et al. 2017; Phan et al.2007; Vipin et al. 2013).

The set of 160 L. albus lines derived from the EuropeanLupin Gene Resources Database maintained by Poznań PlantBreeders Ltd. station located inWiatrowo was used for markervalidation: 79 primitive populations, 36 landraces, 30 culti-vars, 12 cross derivatives, and 3 mutants. These lines originat-ed from 23 countries. Taking into consideration alkaloid con-tent in seeds, 127 lines were bitter (above 0.5% of total dryweight alkaloid content), 26 were sweet (below 0.2%), and 7intermediate (Kroc et al. 2017) (Supplementary File 1).

Plants were grown in a greenhouse at the Institute of PlantGenetics of the Polish Academy of Sciences in Poznań underambient long-day photoperiod (14–16 h). Leaves were col-lected from 4-week plants. DNA was isolated using DNeasyPlant Mini Kit (Qiagen, Hilden, Germany).

Development of PCR-based markers

Marker sequences surrounding pauper locus (Książkiewiczet al. 2017) were aligned to the transcriptome datasets ofKiev and P27174 lines (Książkiewicz et al. 2017) as well asto the reference white lupin gene index LAGI01 (O’Rourkeet al. 2013) by BLAST (Altschul et al. 1990) using Geneioussoftware (Kearse et al. 2012). Matching transcripts were thencomparatively mapped to the genome sequence of the narrow-leafed lupin (Hane et al. 2017), extracting selected loci with10000 nt of flanking regions. To find exon/intron boundaries,

(2019) 60:269–281270 J Appl Genetics

white lupin marker and transcript sequences were assembledtogether with extracted narrow-leafed lupin genome regionsinto contigs using a progressive Mauve algorithm (Darlinget al. 2004) assuming genome collinearity. Mauve alignmentscarrying markers, corresponding white lupin transcripts andfragments of narrow-leafed lupin scaffolds, were searched forthe presence of polymorphic loci. Primers flanking these lociwere designed using Primer3Plus (Untergasser et al. 2007).

PCR was performed using Labcycler thermocycler(SensoQuest GmbH, Göttingen, Germany). GoTaq® G2Flexi DNA Polymerase (Promega) was used. The reactionmixture contained DNA template (2.5 ng/μL), dNTP(0.8 mM), MgCl2 (2 mM), polymerase (1 U), polymerasebuffer (1×), and primers (0.25 μM each).

PCR amplicons were purified directly from the post-reaction mixtures (QIAquick PCR Purification Kit; Qiagen)and sequenced (ABI PRISM 3130 XL Genetic Analyzer;Applied Biosystems, Hitachi) in the Laboratory ofMolecular Biology Techniques, Faculty of Biology, AdamMickiewicz University (Poznan, Poland). Cleaved amplifiedpolymorphic sequence (CAPS) (Konieczny and Ausubel1993) or derived CAPS (dCAPS) (Neff et al. 1998) ap-proaches were used to resolve the nucleotide substitutionpolymorphisms. Restriction sites and dCAPS primers wereidentified using dCAPS Finder 2.0 (Neff et al. 2002) andSNP2CAPS (Thiel et al. 2004). Restriction products wereseparated by agarose gel electrophoresis, with the agaroseconcentration (1–3%) adjusted to follow the size of the ex-pected digestion products.

Linkage mapping

Chi-square (χ2) values for Mendelian segregation in F8 RILswere estimated using the expected 1:1 segregation ratio. Thecalculation of probability was based on χ2 and 2 degrees offreedom. L. albusmarker segregation files (Książkiewicz et al.2017; Phan et al. 2007; Rychel et al. 2019; Vipin et al. 2013),together with those developed in this study, were imported toJoinmap 5.0 (Stam 1993). Mapping procedure was performedas previously described (Książkiewicz et al. 2017). Based onthe initial results of the mapping, the line RIL-169 was re-moved from the final mapping due to the frequent change ofmarker allelic phases. Such an observation may result fromseed admixture or cross-pollination during seed multiplica-tion. Twenty repeats of linkage group calculation with alteredparameters were done to estimate the plausibility of markerpositions. LOD values were calculated in Map ManagerQTXb20 (Manly et al. 2001).

Marker validation

Markers were validated by comparing seed dry weightalkaloid content (Kroc et al. 2017) and marker allelic

phases for the set of 160 collection lines. The Pearsonproduct-moment correlation coefficient was calculated inExcel. Taking into consideration the hypothesis on theinfluence of a single gene (in the pauper locus) on thealkaloid content, binary data similarity analysis was per-formed. Therefore, alkaloid content values above 0.2%were assigned as 1 (bitter and intermediate) and the re-maining values as 0 (sweet), however, known as non-pauper low alkaloid lines were assigned as 1 to avoidunjustified false-negative results. Kiev Mutant-like scoreswere assigned as 0, P27174-like scores as 1, and hetero-zygotes also as 1 because pauper is a recessive allele.Simple matching (Sokal and Michener 1958) andRogers-Tanimoto (Rogers and Tanimoto 1960) coeffi-cients were calculated using binary similarity calculatorhttp://www.minerazzi.com/tools/similari ty/binary-similarity-calculator.php.

Results

Selection of sequences from pauper locus

Molecular markers from the most recent L. albus linkagemap localized in the proximity of the pauper locus werealigned to the L. angustifolius genome and gene sequences(Hane et al. 2017) as well as to the L. albus transcriptome(O’Rourke et al. 2013) to find anchors for primer design.Based on the alignments and positions on the linkage map,6 markers were selected (TP16854, TP22150, TP30216,TP309728, TP447859, TP70046) for further study. Oneof these markers, TP16854, matched Lup021586 gene,which was shown to have 100% nucleotide identity to theLaAT gene (AB581532.1) encoding acyltransferase-likeprotein (Książkiewicz et al. 2017). As the LaAT gene ex-pression was revealed to be correlated with total alkaloidcontent (Bunsupa et al. 2011), it was included in the mark-er array. When the LaAT was blasted against L. albus tran-scriptome, it tagged LAGI01_35805 and LAGI01_49436as the most similar sequences. These transcripts werefound to have high similarity to L. angustifoliusLup021586 and Lup021583 genes, annotated as encodingHXXXD-type acyltransferase family proteins. Analysis ofthe surrounding genes in the L. angustifolius assemblyhighlighted a Lup021589 (a bHLH35-like transcriptionfactor) as a another hypothetical candidate for marker de-velopment. Mapping of the Lup021589 to the L. albustranscriptome resulted in the selection of LAGI01_54458.To summarize, the set of sequences used for pauper markerdevelopment consisted of TP16854, TP22150, TP30216,TP309728, TP447859, TP70046, LAGI01_35805,LAGI01_49436, and LAGI01_54458.

271(2019) 60:269–281Appl GeneticsJ

Development of PCR-based pauper molecular markerarray

For TP16854, TP22150, TP30216, TP309728, TP447859,and TP70046 sequences, markers were developed using thosepolymorphic loci which were published in the linkage mappaper (Ks iążk iewicz e t a l . 2017) . Two dCAPS(TP16854_FD_R and TP70046_F_RD) and four CAPS(TP22150_F_R, TP30216_F_R, TP309728_F_R,TP447859_F_R) markers were designed. Mapping KievMutant and P27174 reads (PRJNA380248) to the L. albustranscriptome revealed polymorphic loci in one exon inLAGI01_49436 and LAGI01_54458, and in two exons inL A G I 0 1 _ 3 5 8 0 5 . T h r e e C A P S m a r k e r s(LAGI01_35805_F1_R1, LAGI01_35805_F2_R2,LAGI01_49436_F2_R2) and one PCR-based INDEL marker(LAGI01_54458_F2_R1) were designed. Initial screening ofwhite lupin lines with published PauperM1 marker (Lin et al.2009) revealed high difficulty of inferring allele genotypesdue to very small difference in product length and amplifica-tion of stutter bands. Therefore, this marker was transformedinto a pair of CAPSmarkers exploiting two different polymor-phic loci. The marker array was supplemented by two gene-based markers, ESD4-F7 and ESD4-F8, which were recentlydeveloped for L. albus homolog of A. thaliana flowering in-duction pathway gene EARLY IN SHORT DAYS 4 (ESD4)and mapped very close to the pauper locus (Rychel et al.2019). The list of developed markers with information onassigned L. angustifolius gene and genome sequences andL. albus transcripts is provided in Table 1. The list of primerpairs, PCR primer annealing temperature, enzyme used forpolymorphism detection, and the lengths of restriction prod-ucts for Kiev Mutant and P27174 lines are provided inTable 2.

Linkage mapping of pauper markers

The segregation of newly developed gene-based markers(LAGI01_49436_F2_R2, LAGI01_35805_F1_R1,LAGI01_35805_F2_R2, LAGI01_54458_F2_R1) was an-alyzed in the RIL population to provide data for linkagemapping. Markers PauperM1, TP16854, TP447859,TP22150, TP309728, TP70046, and TP30216 were al-ready localized on the genetic map (Książkiewicz et al.2017), however, with missing 14–53% of RIL genotypingdata. To increase the quality of linkage mapping, the seg-regation of these markers was tested as well. Segregationdata were obtained for 98.5% of RILs. All markers werelocalized in the linkage group ALB18 in the region carry-ing pauper locus. Both PauperM1 markers revealed iden-tical segregation and localized 1.06 cM upstream thepauper l o cus . F ive ma rke r s (TP16854_FD_R,LAGI01_35805_F2_R2, LAGI01_35805_F1_R1,

LAGI01_49436_F2_R2, and LAGI01_54458_F2_R1) co-segregated with the pauper locus (Fig. 1). The linkage ofthese markers was confirmed by high-LOD values (min49.1, max 56.6, mean 55.25). Markers TP447859_F_R,TP22150_F_R, ESD4-F7, and ESD4-F8 formed a redun-dant cluster localized 0.79 cM downstream the pauper lo-cus. Markers TP309728_F_R, TP70046_F_RD, andTP30216_F_R were mapped at further distances from thepauper, namely 1.06 cM, 2.91 cM, and 7.70 cM.Information on marker χ2 P values for segregation distor-tion, position in linkage group, and LOD values is provid-ed in Table 3. RIL segregation data is given inSupplementary File 2.

Pauper marker validation in collection lines

Validation set of lines contained 127 bitter and five pauper(Kiev Mutant, Boros, Lotos, Hansa, Kali), four exiguus(Start, Nelly USA, Butan, Tombowskij Skorospielyj), onenutricius (Nahrquell), and sixteen other sweet or intermedi-ate accessions with unknown genotype according to the in-formation obtained from the Plant Breeding Smolice Ltd.,Co. (Dr. Stanis ław Stawiński) and published data(Hackbarth 1957; Hackbarth 1961; Harrison and Williams1982; Kroc et al. 2017; Porsche 1964; Šatović 1993; Troll1958). Visualization of marker polymorphism for selectedlines is provided in Supplementary File 3. Simple matchingcoefficients were calculated, to compare the marker geno-type and the pauper phenotype. These values ranged from0.15 to 0.94, indicating rapid linkage disequilibrium decayaround pauper locus. To address the putative applicabilityof newly developed markers in the marker-assisted selec-tion, Rogers-Tanimoto coefficient values were calculated.Rogers-Tanimoto is a variant of the simple matching coef-ficient that gives double weight to mismatching variables,therefore accentuating false-positive and false-negativescores. Rogers-Tanimoto values were in the range from0.08 to 0.88. LAGI54458_F1 marker revealed to have iden-tical simple matching and Rogers-Tanimoto values as thepreviously published PauperM1, namely 0.83 and 0.71.LAGI01_35805_F1_R1marker having those values as highas 0.94 and 0.88 was evidenced to be more applicable tomarker-assisted selection than the PauperM1 (Table 4). Alllines carrying pauper recessive alleles revealed positiveLAGI01_35805_F1_R1 marker score. False-positivescores were obtained for four lines, namely 95015 “SanFelices” (12.73% alkaloid dry weight content), 95064“Population-8062” (2.69%), 95220 “FAM 120” (1.21%),and 95023 “Oeiras-930/3” (4.93%) (Kroc et al. 2017). Allthese lines are primitive accessions or landraces. Markerscores for the set of validation lines are provided inSupplementary File 4.

272 (2019) 60:269–281J Appl Genetics

Table 1 Sequences assigned to developed white lupin markers tagging pauper locus. Positions indicate the first nucleotide in the marker sequencealignment

Marker L. angustifolius chromosome Position (nt) L. angustifolius gene Position (nt) L. albus transcript Position (nt)

ESD4-F7a NLL-16 19 043 934 Lup021600 1 LAGI01_22606 1580

ESD4-F8a NLL-16 19 043 385 Lup021600 417 LAGI01_23187 1003

LAGI01_35805_F1_R1 NLL-16 18 720 613 Lup021586 2 LAGI01_35805 149

LAGI01_35805_F2_R2 NLL-16 18 718 376 Lup021586 231 LAGI01_35805 378

LAGI01_49436_F2_R2 NLL-16 18 680 802 Lup021583 804 LAGI01_49436 71

LAGI01_54458_F2_R1 NLL-16 18 798 603 – – LAGI01_54458 538

PauperM1_F_Rb NLL-16 10 380 713 – – – –

TP16854_FD_R NLL-16 18 718 404 Lup021586 730 LAGI01_35805 880

TP22150_F_R NLL-16 19 032 414 Lup021599 1050 LAGI01_34026 56

TP30216_F_R NLL-16 9 710 799 Lup030539 591 LAGI01_22438 1175

TP309728_F_R NLL-16 19 185 829 Lup021606 773 LAGI01_61485 673

TP447859_F_R NLL-16 19 092 892 Lup021602 389 LAGI01_9900 557

TP70046_F_RD NLL-19 15 265 260 Lup026514 403 LAGI01_45076 417

aMarker published by (Rychel et al. 2019)bMarker published by (Lin et al. 2009)

Table 2 List of developed markers tagging pauper locus, with primer sequences, PCR primer annealing temperature, validated enzymes, andrestriction product sizes

Name Primers PCR (°C) Validated enzyme ProductsKiev (bp)

ProductsP27174 (bp)

TP16854_FD_R CAGCAAGAGATTCAAATGTTGTGAAAGATGAAGGAAATTAGTTGCAATGAATCAA

60 HindIII 64 39, 25

TP447859_F_R CTGCTTCTGATATTTTCAATGCATTAAGCCATCAAATGCTTCAGAAGTAG

60 MnlI 56, 17 73

TP22150_F_R TTGCATGAGGGAAGATGGACGATAGCATGTGACCTGGAGGAC

60 MnlI 81, 48, 25, 8 129, 25, 8

TP309728_F_R TTGAAGGTCTACAAAGGCATCCTATAGACCAGCACAACCCTCTG

60 TaqI 75, 40 115

TP70046_F_RD CCATCATTCCTGCCATAAGGATCAAGGAAGAGTTAGAGAAT

60 HinfI 110, 21 131

TP30216_F_R TATTAAGTCGATTTGGTGGAGCTAATACATGAAATAGGCCCAATGAAA

56 HinfI 76, 34 110

LAGI01_49436_F2_R2 CTCTGAACTTGGGCATGGATGACTTGGTGCCTATGTATGGAGA

64 BamHI 464, 66 527

LAGI01_35805_F1_R1 TGGCATCACTGAAAATTGAGATGACTTTGAGAGGGCTTGTTTGAG

64 BclI 197 108, 89

LAGI01_35805_F2_R2 GCTAGTAAAACAACCCAACGGAGCCCTAGCTCTTGATCTCCAG

64 MboI 253, 216, 109, 16 362, 216, 16

LAGI01_54458_F2_R1 AAAAGTGTTAGAAAAGTACAGCACCAGTGACTCCAACAAAAGCACA

52 – 108 98

ESD4-F8a CTGCGCTTCTCCTGTAGTTCAATTTAGACCTTGAAACCGATAGTCATAATAG

63 MseI 351 178, 173

ESD4-F7a CCCTAACCAACGGTAGCTTGATATTGTGCTTTTTCACCCCTCTC

52 EcoRV 232, 22 254

PauperM1_F_R_HhaIb AAGAAAAGGCCCAATGTTTTAAAGTCATACCATTGAG

54 HhaI ~ 209 108, 99

PauperM1_F_R_HinfIb AAGAAAAGGCCCAATGTTTTAAAGTCATACCATTGAG

54 HinfI 145, ~ 64 207

aMarker published by (Rychel et al. 2019)b Improvement of the PauperM1 marker published by (Lin et al. 2009)

273(2019) 60:269–281Appl GeneticsJ

Discussion

Exploitation of shared synteny for genetic studiesin legumes

In the present study, the synteny-based approach was appliedto design molecular markers delimiting the pauper locus. Theconcept of exploitation of genome collinearity to transfer in-formation from model plants to crop species emerged soonafter sequencing the first three legume species (Cannon et al.2009; Mudge et al. 2005). One of the first examples waspositional cloning of a legume symbiosis LjSym2 gene basedon the comparative mapping between the three genomes dif-fering by the advancement of molecular tools developed,namely Lotus japonicus (linkage map and contigs oftransformation-competent artificial chromosomes), Pisumsativum (linkage map), and A. thaliana (chromosome-scalegenome assembly) (Stracke et al. 2004). The sequence ofA. thaliana genome revealed to be beneficial in mapping can-didate genes conferring important agronomic domestication

traits, growth determination, and photoperiod sensitivity, incommon bean (Kwak et al. 2008). Molecular and comparativemapping combined with classical genetic approach was suc-cessfully applied to decipher the gene underlying yellow/green cotyledon polymorphism, which was first reported byGregor Mendel in 1866 (Armstead et al. 2007). In the earlyyears of legume comparative genomic studies, map-basedcloning strategy resulted in the identification of the RCT1,an M. truncatula resistance gene that confers the multi-raceresistance of alfalfa to a hemibiotrophic fungal pathogenColletotrichum trifolii, causing anthracnose disease (Yanget al. 2008b). Progress in legume genome sequencing provid-ed novel evidence for large-scale synteny existing between thepapilionoid subclades which diverged about 50 million yearsago (Bertioli et al. 2009). When the genome assemblies ofM. truncatula and L. japonicus were aligned to the linkagemap of Phaseolus vulgaris, novel large-scale macrosyntenicblocks were identified, justifying the concept of cross-speciescomparisons for tracking particular domestication genes(McConnell et al. 2010). As an example, a candidate genefor the hypernodulation mutation nod3 in pea was elucidatedby the comparative mapping to M. truncatula genome carry-ing a nodulation regulation Pub1 gene in the syntenic region(Bordat et al. 2011). To facilitate studies involving genomesequence comparisons and synteny-based gene annotation, aLegumeIP 2.0 platform hosting large-scale genomic/transcriptomic data and integrative tools for bioinformaticanalysis has been launched (Li et al. 2012a; Li et al. 2016).The reconstruction of a comparative map composed of sevenspecies from the galegoid clade (M. truncatula, M. sativa,Lens culinaris, P. sativum, L. japonicus, Cicer arietinum,Vicia faba) and three species from the phaseoloid clade(Vigna radiata, P. vulgaris, Glycine max) carrying cross-species gene-derived markers revealed numerousmacrosyntenic segments shared between all species analyzed(Lee et al. 2017). Recently, ten sequenced legume genomeswere hierarchically aligned to establish a family-level geno-mics platform for studying evolutionary changes as well asfunctional analysis of genes involved in regulatory pathways(Wang et al. 2017).

Among lupins, L. angustifolius was the first speciessubjected to genetic map development and physical map-ping studies (Boersma et al. 2005; Kaczmarek et al. 2009;Kasprzak et al. 2006; Kruszka and Wolko 1999;Leśniewska et al. 2011; Nelson et al. 2006). Comparativemapping of the progressively improved versions ofL. angustifolius linkage map to sequence legume genomesrevealed multiple blocks of conserved synteny carryinggene-rich regions and some candidate domestication genes(Kamphuis et al. 2015; Kroc et al. 2014; Książkiewiczet al. 2016; Książkiewicz et al. 2013; Książkiewicz et al.2015; Nelson et al. 2010; Nelson et al. 2006; Przysieckaet al. 2015; Wyrwa et al. 2016). The synteny-based

Fig. 1 Localization of markers in the white lupin linkage group ALB18region carrying pauper locus. Vertical bar graph visualizes linkage groupfragment with cumulative genetic distance scale expressed incentimorgans (cM). The linear plot shows corresponding LOD valuesof linkage to adjacent markers. Names of markers analyzed in thisstudy are in boldface

274 (2019) 60:269–281J Appl Genetics

approach was further exploited to identify the particulargene (LanFTc1, a homolog of A. thaliana FLOWERINGLOCUS T) underlying one of the major domestication lociin L. angustifolius, Ku, conferring vernalization indepen-dence and early flowering (Nelson et al. 2017; Nelson et al.2006). Even the mechanism of regulation, based on therelatively long insertion/deletions in the promoter se-quence was revealed to be conserved (Liu et al. 2014;Taylor et al. 2019). The pattern of cross-species synteny

facilitated also the assembly of the L. angustifoliuspseudochromosomes, providing anchors for scaffold orderand orientation in the regions where the marker resolutionwas insufficient due to low recombination rate (Hane et al.2017). This assembly has been further improved with theaid of an ultra-high-density genetic map containing 34574sequence-defined markers (Zhou et al. 2018).

Molecular studies on white lupin based on comparativemapping approaches have been hampered for many years

Table 3 RIL segregation data, chi-square P values for segregation distortion, linkage group positions, and LOD scores to adjacent loci calculated forwhite lupin pauper markers

Name % RIL data χ2

P valueLinkage group Locus (cM) LOD scores

PauperM1_F_R_HhaI 99.0 0.75 ALB18 68.57 52.7, 58.1

PauperM1_F_R_HinfI 99.0 0.75 ALB18 68.57 58.1, 49.1

paupera 95.4 0.66 ALB18 69.64 55.4, 54.8

LAGI01_49436_F2_R2 96.4 0.76 ALB18 69.64 56.3, 56.3

TP16854_FD_R 99.0 0.83 ALB18 69.64 49.1, 55.4

LAGI01_35805_F2_R2 97.4 0.78 ALB18 69.64 54.8, 56.6

LAGI01_35805_F1_R1 98.0 0.71 ALB18 69.64 56.6, 56.3

LAGI01_54458_F2_R1 98.0 0.83 ALB18 69.64 56.3, 54.8

TP22150_F_R 99.5 0.78 ALB18 70.43 58.1, 58.4

ESD4-F7 99.5 0.78 ALB18 70.43 58.4, 58.4

ESD4-F8 99.5 0.78 ALB18 70.43 58.4, 55.1

TP447859_F_R 99.0 0.83 ALB18 70.43 54.8, 58.1

TP309728_F_R 98.5 0.78 ALB18 70.69 55.1, 44.2

TP70046_F_RD 96.4 0.83 ALB18 72.55 44.2, 33.6

TP30216_F_R 99.5 0.25 ALB18 77.34 33.6, 16.0

a Locus published by (Książkiewicz et al. 2017; Phan et al. 2007)

Table 4 Validation of pauper markers in white lupin collection lines differing in alkaloid content

Marker Simple matching coefficientwith bitter/sweet score

Rogers-Tanimoto coefficientwith bitter/sweet score

Correlation withbitter/sweet score

Correlation with %alkaloid content

LAGI01_35805_F1_R1 0.94 0.88 0.71 0.26

LAGI01_54458_F2_R1 0.83 0.71 0.41 0.15

PauperM1_F_R_HhaI 0.83 0.71 0.45 0.15

PauperM1_F_R_HinfI 0.83 0.71 0.45 0.15

TP70046_F_RD 0.68 0.52 0.33 0.27

TP16854_FD_R 0.64 0.47 0.34 0.31

TP30216_F_R 0.52 0.35 0.18 0.09

TP309728_F_R 0.49 0.33 0.09 0.02

TP447859_F_R 0.47 0.30 0.07 0.06

LAGI01_35805_F2_R2 0.45 0.29 0.25 0.26

LAGI01_49436_F2_R2 0.31 0.18 0.17 0.16

TP22150_F_R 0.18 0.10 0.03 − 0.05ESD4-F8 0.15 0.08 0.05 − 0.01

ESD4-F7 0.15 0.08 0.05 − 0.01

275(2019) 60:269–281Appl GeneticsJ

because the published linkage maps carried a relativelysmall number of sequenced markers (Croxford et al.2008; Phan et al. 2007; Vipin et al. 2013). The most recentlinkage map of white lupin with 3669 sequenced markershighlighted the collinearity between L. angustifolius andL. albus genomes and provided novel possibilities formap-based gene cloning (Książkiewicz et al. 2017). Thistool was further exploited to identify candidate genes in-volved in white lupin early flowering (Rychel et al. 2019).The information on highly conserved synteny betweenL. albus and L. angustifolius genomes was also harnessedin the present study to design new markers tagging lowalkaloid pauper locus.

Markers and genes for low alkaloid content in lupins

Alkaloid content as a major potentially toxic anti-nutritionalfactor was thoroughly investigated during the lupin domesti-cation process, and numerous low alkaloid accessions wereselected in all three crop Old World lupin species:L. angustifolius, L. albus, and L. luteus (Hackbarth 1957;Hackbarth 1961; Hackbarth and Troll 1956; Święcicki 1986;Święcicki and Jach 1980; Święcicki and Święcicki 1995).Significant progress in the determination of low alkaloid lineshas been also achieved in the main New World lupin crop,L. mutabil is (Galek et al . 2017). Among lupins,L. angustifolius has been subjected to the most advanced stud-ies on genetic and molecular factors affecting quinolizidinealkaloid biosynthesis. Three major recessive low-alkaloid al-leles were identified in L. angustifolius germplasm, namelyiucundus, depressus, and esculentus; however, only iucunduswas widely introduced into breeding programs (Święcicki andŚwięcicki 1995). L. angustifolius RIL mapping populationdeveloped from the cross of 83A:476 (maternal, sweet, do-mesticated) and P27255 (paternal, bitter, wild) enabled genet-ic localization of iucundus locus. However, this trait revealedhigh distortion from the expected 1:1 segregation ratio, evi-denced by the chi-square P value of 0.008 (Boersma et al.2005). Iucundus was localized in all versions of theL. angustifolius linkage map but was surrounded only byMFLP-derived markers lacking sequence information(Boersma et al. 2005; Kamphuis et al. 2015; Nelson et al.2010). To provide a DNA marker tightly linked to iucundus(~ 0.9 cM), a separate study was performed involving 20 linesand 320 MFLP fingerprints (Li et al. 2011). Novel markersclosely related to the iucunduswere developed during genomesequencing attempts; however, no candidate genewas hypoth-esized (Hane et al. 2017; Zhou et al. 2018). Recently,transcriptome-based studies revealed high correlation of alka-loid content with leaf tissue expression levels of genesencoding lysine/ornithine decarboxylase (LaL/ODC), copperamine oxidase (LaCAO), acyltransferase (LaAT), berberinebridge enzyme (LaBBE-like), and major latex-like proteins

(LaMLP1-like, LaMLP2-like, and LAMLP4-like) (Frick et al.2018; Yang et al. 2017). The group of L. angustifoliusquinolizidine alkaloid biosynthesis genes is putatively regu-lated by an APETALA2/ethylene responsive transcription fac-tor, which was evidenced by linkage mapping and tran-scriptome profiling as a strong candidate for iucundus (Krocet al. 2019).

In L. luteus, four low-alkaloid alleles were identified, in-cluding dulcis, amoenus, liber (von Sengbusch 1942), and v(Gustafsson and Gadd 1965). Forms with alkaloid contentbelow 0.05% were developed (Święcicki and Jach 1980).Molecular resources for this species are very limited and in-clude two transcriptome assemblies derived from independentstudies and the set of insertion/deletion markers developed bynext-generation sequencing of genomic reduction libraries(Glazinska et al. 2017; Osorio et al. 2018; Parra-Gonzálezet al. 2012). The lack of mapping populations and linkagemaps considerably impeded research on genes underlyinglow alkaloid content in this species.

Genetic studies involving L. albus germplasm resulted inthe identification of several loci underlying low alkaloid con-tent: pauper/primus/tercius, exiguus, nutricius, mitis, suavis,reductus, and minutus (Hackbarth 1957; Hackbarth 1961;Harrison and Williams 1982; Porsche 1964; Šatović 1993;Troll 1958). Pauper was widely exploited for breeding;exiguus and nutricius were used occasionally, whereas otherloci remained untapped (Harrison andWilliams 1982; Šatović1993). First two linkage maps addressing pauper segregationdid not provide any marker closely related to this gene (Phanet al. 2007; Vipin et al. 2013). With the aid of the MFLPtechnique, a PCR-based PauperM1 marker tagging pauperlocus by 1.4 cM was developed (Lin et al. 2009). However,due to the small length difference between the alleles and theproduction of some background stutter bands, analysis of thismarker required tedious and time-consuming sequencing gelelectrophoresis.

A step towards identification of the gene underlyingpauper locus

In this paper, the PauperM1 marker was improved to a CAPSmarker addressing two closely located SNPs, recognized bytwo different enzymes. Enzyme HhaI is expected to cut thebitter allele, whereas HinfI—the sweet one. Such an approachminimizes the risk of false-positive and false-negative scoresresulting from non-occurrence of cleavage due to reactionpreparation issues. In the present study, six SNP markers gen-erated by genotyping-by-sequencing (Książkiewicz et al.2017) were transformed to PCR-based markers using CAPSand dCAPS approaches (Konieczny and Ausubel 1993; Neffet al. 1998). It is widely adapted strategy for scoring SNPmarkers obtained by high-throughput sequencing(Shavrukov 2016). Several tools were developed, allowing

276 (2019) 60:269–281J Appl Genetics

design of CAPS and/or dCAPS markers on one-by-one basis(dCAPS Finder, BlastDigester, SNP2CAPS, SGN CAPSDesigner) or as a high-throughput automated process(CAPS/dCAPS Designer) (Ilic et al. 2004; Li et al. 2018;Neff et al. 2002; Thiel et al. 2004). For routine implementa-tion, other PCR allelic discrimination technologies are consid-ered, including rhAmp, TaqMan, or KASP assays(Broccanello et al. 2018). Indeed, a Fluidigm nanofluidic ar-ray genotyping platform has been exploited to formulateL. angustifolius SNP array and provide markers for linkagemapping and genome assembly (Hane et al. 2017; Kamphuiset al. 2015; Yang et al. 2013; Zhou et al. 2018).

O n e o f t h e n e w l y d e v e l o p e d m a r k e r s ,LAGI01_35805_F1_R1, was revealed to have higher applica-bility for paupermarker-assisted selection than the previouslypublished PauperM1 marker (Lin et al. 2009), evidenced byhigher values of all coefficients calculated (simple matching,Rogers-Tonimoto, Pearson product-moment correlation).However, false-positive scores were revealed for four lines(constituting 2.5% of analyzed plant materials). All these lineswere primitive accessions or landraces, including a line withthe highest dry weight seed alkaloid content in the collection,95015 “San Felices” (Kroc et al. 2017). Such an observationmay indicate that the pauper locus gene is different than thegene represented by the LAGI01_35805_F1_R1 marker se-quence. It is also possible that the LAGI01_35805 is derivedfrom the true pauper locus gene but the SNP recognized bythis marker is not the functional mutation causing low alkaloidcontent. LAGI01_35805 is a homolog of L. angustifoliusLup021586 gene , anno ta t ed as the LaAT gene(AB581532.1) (Książkiewicz et al. 2017). LaAT is a represen-tative of BAHD acyl-CoA-dependent acyltransferase super-family and was shown to be highly expressed in the leavesof quinolizidine alkaloid-producing L. angustifolius plants butundetectable in the sweet ones (Bunsupa et al. 2011).However, in L. angustifolius, the function of low alkaloidiucundus gene is assigned rather to a regulatory agent (a tran-scription factor) than to an enzyme directly involved in alka-loid biosynthesis (Kroc et al. 2019). Nevertheless,L. angustifolius and L. albus have a relatively different patternof alkaloid compound variation and partially differ by majorcomponent influencing total alkaloid content (Boschin et al.2008; Kamel et al. 2016; Kroc et al. 2017). It was suggestedthat functions of iucundus and pauper genes may be distinctbecause these species have different lysine profiles amongwild and sweet accessions (Frick et al. 2017). As expected,negative LAGI01_35805_F1_R1 scores were obtained forfour low alkaloid lines carrying exiguus gene (95422 “Start”,95480 “Nelly”, 95513 “Butan” derived from Start × Wat,95454 “Tombowskij Skorospielyj”) and one carryingnutricius gene (95509 “Nahrquell”) (Hackbarth 1957;Hackbarth 1961; Harrison and Williams 1982; Porsche1964; Šatović 1993; Troll 1958; Stawiński S. unpublished).

Additionally, negative LAGI01_35805_F1_R1 scores wererevealed for six sweet lines with unknown low-alkaloid donor(95175 “R-84141”, 95433 “Dniepr”, 95160 “R-243”, 95166“SF-479”, 95445 “Lutrop,” and 95449 “Buttercup”). Theselines may carry exiguus, nutricius, or other low-alkaloidgenes.

In the sister crop species, narrow-leafed lupin, the breed-ing process has been considerably facilitated by markerswhich were developed to select key agronomic traits andsubsequently implemented in Australian breeding pro-grams. These include ear ly f lower ing (KuHM1,LanFTc1_INDEL) (Boersma et al. 2007a; Nelson et al.2017), reduced pod shattering (TaLi, TaM1, TaM2, LeM1,LeM2, LeLi) (Boersma et al. 2007b; Boersma et al. 2009; Liet al. 2010; Li et al. 2012c), low alkaloid profile (iucLi) (Liet al. 2011), soft seediness (marker MoLi) (Li et al. 2012b),and resistance to diseases caused by pathogenic fungi, in-cluding anthracnose (AntjM1, AntjM2, AnManM1) (Yanget al. 2004; Yang et al. 2008a; You et al. 2005) andPhomopsis stem blight (PhtjM1, PhtjM2, Ph258M1,Ph258M2) (Yang et al. 2002). Numerous studies revealedthat the further improvement of white lupin as a crop willrequire incorporation of rare alleles, such as resistance toanthracnose found only in Ethiopian lines, from wild land-races which are bitter and late flowering (Adhikari et al.2009; Adhikari et al. 2013; Phan et al. 2007). Reselectionof agronomic traits in the progeny could be greatly facilitat-ed by the use of markers targeting particular domesticationgenes. Such a model was established for the narrow-leafedlupin (Cowling et al. 2009). Markers for low alkaloidpauper locus developed in this study, together with thoserecently published for early flowering (Rychel et al. 2019),address this requirement and constitute a versatile array forwhite lupin molecular breeding.

Acknowledgments We thank Prof. Paolo Annicchiarico and Dr. NelsonNazzicari from CREA-FLC, Council for Agricultural Research andEconomics, Research Centre for Fodder Crops and Dairy Production(Lodi, Italy), for early access to L. albus genotyping-by-sequencing data(Książkiewicz et al. 2017). We thank the Department of Agriculture andFood ofWestern Australia (Perth, Australia) for the seeds of KievMutant× P27174 mapping population. We thank Poznan Plant Breeding Ltd.station located in Wiatrowo for the seeds of white lupin lines for paupermarker genotyping assay. We also thank Dr. Matthew Nelson from theCommonwealth Science and Industry Research Organisation (Wembley,Western Australia, Australia) for pre-publication access to the white lupintranscriptome data (SRX2663946 and SRX2663947).

Author’s contribution SR performed all experimental procedures (plantcultivation, DNA isolation, PCR, restriction enzyme cleavage, agarosegel electrophoresis) and prepared result files with genotyping data. MKdesigned molecular markers, performed linkage mapping and other cal-culations, and drafted the manuscript.

Funding This research was funded by the European Community’sSeventh Framework Programme LEGATO project (FP7-613551).

277(2019) 60:269–281Appl GeneticsJ

Compliance with ethical standards

This article does not contain any studies with human participants or an-imals, performed by any of the authors.

Conflict of interest The authors declare that they have no conflict ofinterest.

Data availability All data generated during this study are included inthis published article and its supplementary information files. Markersequences were deposited in the DNA Data Bank of Japan (accessionnumbers LC466085-LC466106).

Open Access This article is distributed under the terms of the CreativeCommons At t r ibut ion 4 .0 In te rna t ional License (h t tp : / /creativecommons.org/licenses/by/4.0/), which permits unrestricted use,distribution, and reproduction in any medium, provided you giveappropriate credit to the original author(s) and the source, provide a linkto the Creative Commons license, and indicate if changes were made.

References

Adhikari KN, Buirchell BJ, Thomas GJ, Sweetingham MW, Yang H(2009) Identification of anthracnose resistance in Lupinus albus L.and its transfer from landraces to modern cultivars. Crop Pasture Sci60:472–479. https://doi.org/10.1071/CP08092

Adhikari KN, Thomas G, Diepeveen D, Trethowan R (2013)Overcoming the barriers of combining early flowering and anthrac-nose resistance in white lupin (Lupinus albus L.) for the NorthernAgricultural Region ofWestern Australia. Crop Pasture Sci 64:914–921. https://doi.org/10.1071/CP13249

Altschul SF, GishW,MillerW,Myers EW, Lipman DJ (1990) Basic localalignment search tool. J Mol Biol 215:403–410. https://doi.org/10.1016/S0022-2836(05)80360-2

Armstead I et al (2007) Cross-species identification of Mendel’s I locus.Science 315:73. https://doi.org/10.1126/science.1132912

Arnoldi A, Greco S (2011) Nutritional and nutraceutical characteristics oflupin protein. Nutrafoods 10:23–29. https://doi.org/10.1007/bf03223356

Bertioli DJ et al (2009) An analysis of synteny of Arachis with Lotus andMedicago sheds new light on the structure, stability and evolution oflegume genomes. BMC Genomics 10:45. https://doi.org/10.1186/1471-2164-10-45

Boersma JG, Buirchell BJ, Sivasithamparam K, Yang H (2007a)Development of a sequence-specific PCR marker linked to the Kugene which removes the vernalization requirement in narrow-leafedlupin. Plant Breed 126:306–309. https://doi.org/10.1111/j.1439-0523.2007.01347.x

Boersma JG, Buirchell BJ, Sivasithamparam K, Yang H (2007b)Development of two sequence-specific PCR markers linked to thele gene that reduces pod shattering in narrow-leafed lupin (Lupinusangustifolius L). Genet Mol Biol 30:623–629. https://doi.org/10.1590/S1415-47572007000400020

Boersma JG, Nelson MN, Sivasithamparam K, Yang Ha (2009)Development of sequence-specific PCR markers linked to theTardus gene that reduces pod shattering in narrow-leafed lupin(Lupinus angustifolius L). Mol Breed 23:259–267. https://doi.org/10.1007/s11032-008-9230-2

Boersma JG, Pallotta M, Li C, Buirchell BJ, Sivasithamparam K, Yang H(2005) Construction of a genetic linkage map using MFLP andidentification of molecular markers linked to domestication genes

in narrow-leafed lupin (Lupinus angustifolius L). Cell Mol Biol Lett10:331–344

Bordat A et al (2011) Translational genomics in legumes allowed placingin silico 5460 Unigenes on the pea functional map and identifiedcandidate genes in Pisum sativum L. G3: Genes|Genomes|Genetics1:93–103. https://doi.org/10.1534/g3.111.000349

Boschin G, Annicchiarico P, Resta D, D'Agostina A, Arnoldi A (2008)Quinolizidine alkaloids in seeds of lupin genotypes of different or-igins. J Agric Food Chem 56:3657–3663. https://doi.org/10.1021/jf7037218

Boschin G, D’Agostina A, Annicchiarico P, Arnoldi A (2007) The fattyacid composition of the oil from Lupinus albus cv. Luxe as affectedby environmental and agricultural factors. Eur Food Res Technol225:769–776. https://doi.org/10.1007/s00217-006-0480-0

Broccanello C, Chiodi C, Funk A, McGrath JM, Panella L, Stevanato P(2018) Comparison of three PCR-based assays for SNP genotypingin plants. Plant Methods 14:28. https://doi.org/10.1186/s13007-018-0295-6

Bunsupa S, Okada T, Saito K, Yamazaki M (2011) An acyltransferase-like gene obtained by differential gene expression profiles ofquinolizidine alkaloid-producing and nonproducing cultivars ofLupinus angustifolius. Plant Biotechnol 28:89–94. https://doi.org/10.5511/plantbiotechnology.10.1109b

Cannon SB, May GD, Jackson SA (2009) Three sequenced legume ge-nomes and many crop species: rich opportunities for translationalgenomics. Plant Physiol 151:970–977. https://doi.org/10.1104/pp.109.144659

Cowling WA, Buirchell BJ, Falk DE (2009) A model for incorporatingnovel alleles from the primary gene pool into elite crop breedingprograms while reselecting major genes for domestication or adap-tation. Crop Pasture Sci 60:1009–1015. https://doi.org/10.1071/CP08223

Croxford AE, Rogers T, Caligari PDS, Wilkinson MJ (2008) High-resolution melt analysis to identify and map sequence-tagged siteanchor points onto linkage maps: a white lupin (Lupinus albus) mapas an exemplar. New Phytol 180:594–607. https://doi.org/10.1111/j.1469-8137.2008.02588.x

Darling AC, Mau B, Blattner FR, Perna NT (2004) Mauve: multiplealignment of conserved genomic sequence with rearrangements.Genome Res 14:1394–1403. https://doi.org/10.1101/gr.2289704

Daverio M, Cavicchiolo ME, Grotto P, Lonati D, Cananzi M, Da Dalt L(2014) Bitter lupine beans ingestion in a child: a disregarded causeof acute anticholinergic toxicity. Eur J Pediatr 173:1549–1551.https://doi.org/10.1007/s00431-013-2088-2

Frick KM, Foley RC, Kamphuis LG, Siddique KHM, Garg G, Singh KB(2018) Characterization of the genetic factors affecting quinolizidinealkaloid biosynthesis and its response to abiotic stress in narrow-leafed lupin (Lupinus angustifolius L.). Plant Cell Environ 41:2155–2168. https://doi.org/10.1111/pce.13172

Frick KM, Kamphuis LG, Siddique KH, Singh KB, Foley RC (2017)Quinolizidine alkaloid biosynthesis in lupins and prospects for grainquality improvement. Front Plant Sci 8:87. https://doi.org/10.3389/fpls.2017.00087

Galek R, Sawicka-Sienkiewicz E, Zalewski D, Stawiński S, Spychała K(2017) Searching for low alkaloid forms in the Andean lupin(Lupinus mutabilis) collection. Czech J Genet Plant Breed 53:55–62. https://doi.org/10.17221/71/2016-CJGPB

Gladstones JS (1970) Lupins as crop plants. Field Crop Abstracts 23:26Glazinska P et al (2017) De novo transcriptome profiling of flowers,

flower pedicels and pods of Lupinus luteus (yellow lupine) revealscomplex expression changes during organ abscission. Front PlantSci 8:641. https://doi.org/10.3389/fpls.2017.00641

Gustafsson A, Gadd I (1965) Mutations and crop improvement. II Thegenus Lupinus (Leguminosae). Hereditas 53:15–39

Hackbarth J (1957) Die Gene der Lupinenarten. III Weiße Lupine(Lupinus albus). Zeitschrift für Pflanzenzüchtung 37:185–191

278 (2019) 60:269–281J Appl Genetics

Hackbarth J (1961) Untersuchungen über die Vererbung derAlkaloidarmut bei der Weiss-lupine (Lupinus albus). Zeitschrift fürPflanzenzüchtung 45:334–344

Hackbarth J, Troll HJ (1956) Die Lupinen als Körnerleguminosen undFutterpflanzen. In: Handbuch der Pflanzenzüchtung. Verlag PaulParey, Berlin, pp 1–51

Hane JK et al (2017) A comprehensive draft genome sequence for lupin(Lupinus angustifolius), an emerging health food: insights intoplant-microbe interactions and legume evolution. Plant BiotechnolJ 15:318–330. https://doi.org/10.1111/pbi.12615

Harrison JEM, Williams W (1982) Genetical control of alkaloids inLupinus albus. Euphytica 31:357–364. https://doi.org/10.1007/bf00021651

Ilic K, Berleth T, Provart NJ (2004) BlastDigester-a web-based programfor efficient CAPS marker design. Trends Genet 20:280–283.https://doi.org/10.1016/j.tig.2004.04.012

Kaczmarek A, Naganowska B, Wolko B (2009) Karyotyping of thenarrow-leafed lupin (Lupinus angustifolius L.) by using FISH,PRINS and computer measurements of chromosomes. J ApplGenet 50:77–82. https://doi.org/10.1007/BF03195657

Kamel KA, Święcicki W, Kaczmarek Z, Barzyk P (2016) Quantitativeand qualitative content of alkaloids in seeds of a narrow-leafed lupin(Lupinus angustifolius L.) collection. Genet Resour Crop Evol 63:711–719. https://doi.org/10.1007/s10722-015-0278-7

Kamphuis LG, Hane JK, Nelson MN, Gao L, Atkins CA, Singh KB(2015) Transcriptome sequencing of different narrow-leafed lupintissue types provides a comprehensive uni-gene assembly and ex-tensive gene-basedmolecular markers. Plant Biotechnol J 13:14–25.https://doi.org/10.1111/pbi.12229

Kasprzak A, Safár J, Janda J, Dolezel J, Wolko B, Naganowska B (2006)The bacterial artificial chromosome (BAC) library of the narrow-leafed lupin (Lupinus angustifolius L). Cell Mol Biol Lett 11:396–407. https://doi.org/10.2478/s11658-006-0033-3

Kearse M et al (2012) Geneious basic: an integrated and extendabledesktop software platform for the organization and analysis of se-quence data. Bioinformatics 28:1647–1649. https://doi.org/10.1093/bioinformatics/bts199

Konieczny A, Ausubel FM (1993) A procedure for mapping Arabidopsismutations using co-dominant ecotype-specific PCR-based markers.Plant J 4:403–410. https://doi.org/10.1046/j.1365-313X.1993.04020403.x

Kroc M et al (2019) Transcriptome-derived investigation of biosynthesisof quinolizidine alkaloids in narrow-leafed lupin (Lupinusangustifolius L.) highlights candidate genes linked to iucundus lo-cus. Sci Rep 9:2231. https://doi.org/10.1038/s41598-018-37701-5

Kroc M, Koczyk G, Święcicki W, Kilian A, Nelson MN (2014) Newevidence of ancestral polyploidy in the Genistoid legume Lupinusangustifolius L. (narrow-leafed lupin). Theor Appl Genet 127:1237–1249. https://doi.org/10.1007/s00122-014-2294-y

Kroc M, Rybiński W, Wilczura P, Kamel K, Kaczmarek Z, Barzyk P,Święcicki W (2017) Quantitative and qualitative analysis of alka-loids composition in the seeds of a white lupin (Lupinus albus L.)collection. Genet Resour Crop Evol 64:1853–1860. https://doi.org/10.1007/s10722-016-0473-1

Kruszka K, Wolko B (1999) Linkage maps of morphological and molec-ular markers in lupin. In: van Santen E, Wink M, Weissmann S,Roemer P (eds) 9th international lupin conference, Klink/Müritz,Germany, 1999. International Lupin Association, Canterbury, NewZealand, pp 100–106

Książkiewicz M et al. (2017) A high-density consensus linkage map ofwhite lupin highlights synteny with narrow-leafed lupin and pro-vides markers tagging key agronomic traits. Sci Rep 7:15335.https://doi.org/10.1038/s41598-017-15625-w

Książkiewicz M, Rychel S, Nelson MN, Wyrwa K, Naganowska B,Wolko B (2016) Expansion of the phosphatidylethanolaminebinding protein family in legumes: a case study of Lupinus

angustifolius L. FLOWERING LOCUS T homologs, LanFTc1and LanFTc2. BMC Genomics 17:820. https://doi.org/10.1186/s12864-016-3150-z

Książkiewicz M et al (2013) Comparative genomics of Lupinusangustifolius gene-rich regions: BAC library exploration, geneticmapping and cytogenetics. BMC Genomics 14:79. https://doi.org/10.1186/1471-2164-14-79

Książkiewicz M et al (2015) Remnants of the legume ancestral genomepreserved in gene-rich regions: insights from Lupinus angustifoliusphysical, genetic, and comparative mapping. Plant Mol Biol Report33:84–101. https://doi.org/10.1007/s11105-014-0730-4

KwakM, Velasco D, Gepts P (2008)Mapping homologous sequences fordeterminacy and photoperiod sensitivity in common bean(Phaseolus vulgaris). J Hered 99:283–291. https://doi.org/10.1093/jhered/esn005

Lambers H, Clements JC, Nelson MN (2013) How a phosphorus-acquisition strategy based on carboxylate exudation powers the suc-cess and agronomic potential of lupines (Lupinus, Fabaceae). Am JBot 100:263–288. https://doi.org/10.3732/ajb.1200474

Lee C, Yu D, Choi H-K, Kim RW (2017) Reconstruction of a compositecomparative map composed of ten legume genomes. Genes & ge-nomics 39:111–119. https://doi.org/10.1007/s13258-016-0481-8

Leśniewska K, Książkiewicz M, Nelson MN, Mahé F, Aïnouche A,Wolko B, Naganowska B (2011) Assignment of 3 genetic linkagegroups to 3 chromosomes of narrow-leafed lupin. J Hered 102:228–236. https://doi.org/10.1093/jhered/esq107

Li J, Dai X, Liu T, Zhao PX (2012a) LegumeIP: an integrative databasefor comparative genomics and transcriptomics of model legumes.Nucleic Acids Res 40:D1221–D1229. https://doi.org/10.1093/nar/gkr939

Li J, Dai X, Zhuang Z, Zhao PX (2016) LegumeIP 2.0–a platform for thestudy of gene function and genome evolution in legumes. NucleicAcids Res 44:D1189–D1194. https://doi.org/10.1093/nar/gkv1237

Li L, Liu J, Xue X, Li C, Yang Z, Li T (2018) CAPS/dCAPS designer: aweb-based high-throughput dCAPS marker design tool. Sci ChinaLife Sci 61:992–995. https://doi.org/10.1007/s11427-017-9286-y

Li X, Buirchell B, Yan G, Yang H (2012b) A molecular marker linked tothe mollis gene conferring soft-seediness for marker-assisted selec-tion applicable to a wide range of crosses in lupin (Lupinusangustifolius L.) breeding. Mol Breed 29:361–370. https://doi.org/10.1007/s11032-011-9552-3

Li X, Renshaw D, Yang H, Yan G (2010) Development of a co-dominantDNA marker tightly linked to gene tardus conferring reduced podshattering in narrow-leafed lupin (Lupinus angustifolius L).Euphytica 176:49–58. https://doi.org/10.1007/s10681-010-0212-1

Li X, Yang H, Buirchell B, Yan G (2011) Development of a DNAmarkertightly linked to low-alkaloid gene iucundus in narrow-leafed lupin(Lupinus angustifolius L.) for marker-assisted selection. CropPasture Sci 62:218–224. https://doi.org/10.1071/CP10352

Li X, Yang H, Yan G (2012c) Development of a co-dominant DNAmarker linked to the gene lentus conferring reduced pod shatteringfor marker-assisted selection in narrow-leafed lupin (Lupinusangustifolius) breeding. Plant Breed 131:540–544. https://doi.org/10.1111/j.1439-0523.2012.01978.x

Lin R et al (2009) Development of a sequence-specific PCR markerlinked to the gene “pauper” conferring low-alkaloids in white lupin(Lupinus albus L.) for marker assisted selection. Mol Breed 23:153–161. https://doi.org/10.1007/s11032-008-9222-2

Liu L, Adrian J, Pankin A, Hu J, Dong X, von Korff M, Turck F (2014)Induced and natural variation of promoter length modulates the pho-toperiodic response of FLOWERING LOCUS T. Nat Commun 5:4558. https://doi.org/10.1038/ncomms5558

Manly KF, Robert H, Cudmore J, Meer JM (2001) Map Manager QTX,cross-platform software for genetic mapping. Mamm Genome 12:930–932. https://doi.org/10.1007/s00335-001-1016-3

279(2019) 60:269–281Appl GeneticsJ

McConnellM,Mamidi S, Lee R, Chikara S, RossiM, Papa R,McClean P(2010) Syntenic relationships among legumes revealed using agene-based genetic linkage map of common bean (Phaseolusvulgaris L). Theor Appl Genet 121:1103–1116. https://doi.org/10.1007/s00122-010-1375-9

Mohamed AA, Rayas-Duarte P (1995) Composition of Lupinus albus.Cereal Chem 72:643–647

Mudge J, Cannon SB, Kalo P, Oldroyd GE, Roe BA, Town CD, YoungND (2005) Highly syntenic regions in the genomes of soybean,Medicago truncatula, and Arabidopsis thaliana. BMC Plant Biol5:15. https://doi.org/10.1186/1471-2229-5-15

Neff MM, Neff JD, Chory J, Pepper AE (1998) dCAPS, a simple tech-nique for the genetic analysis of single nucleotide polymorphisms:experimental applications in Arabidopsis thaliana genetics. Plant J14:387–392. https://doi.org/10.1046/j.1365-313X.1998.00124.x

Neff MM, Turk E, Kalishman M (2002) Web-based primer design forsingle nucleotide polymorphism analysis. Trends Genet 18:613–615. https://doi.org/10.1016/S0168-9525(02)02820-2

Nelson MN et al (2017) The loss of vernalization requirement in narrow-leafed lupin is associated with a deletion in the promoter and de-repressed expression of a Flowering Locus T (FT) homologue. NewPhytol 213:220–232. https://doi.org/10.1111/nph.14094

NelsonMN et al (2010) Aligning a new reference genetic map of Lupinusangustifoliuswith the genome sequence of the model legume, Lotusjaponicus. DNA Res 17:73–83. https://doi.org/10.1093/dnares/dsq001

Nelson MN et al (2006) The first gene-based map of Lupinusangustifolius L.-location of domestication genes and conservedsynteny with Medicago truncatula. Theor Appl Genet 113:225–238. https://doi.org/10.1007/s00122-006-0288-0

O'Rourke JA et al (2013) An RNA-Seq transcriptome analysis oforthophosphate-deficient white lupin reveals novel insights intophosphorus acclimation in plants. Plant Physiol 161:705–724.https://doi.org/10.1104/pp.112.209254

Osorio CE, Udall JA, Salvo-Garrido H, Maureira-Butler IJ (2018)Development and characterization of InDel markers for Lupinusluteus L. (Fabaceae) and cross-species amplification in other Lupinspecies. Electron J Biotechnol 31:44–47. https://doi.org/10.1016/j.ejbt.2017.11.002

Papineau J, Huyghe C (2004) Le lupin doux protéagineux. Franceagricole, Paris

Parra-González LB et al (2012) Yellow lupin (Lupinus luteus L.) tran-scriptome sequencing: molecular marker development and compar-ative studies. BMCGenomics 13:425. https://doi.org/10.1186/1471-2164-13-425

Phan HTT, Ellwood SR, Adhikari K, Nelson MN, Oliver RP (2007) Thefirst genetic and comparative map of white lupin (Lupinus albus L.):identification of QTLs for anthracnose resistance and floweringtime, and a locus for alkaloid content. DNA Res 14:59–70. https://doi.org/10.1093/dnares/dsm009

Porsche W (1964) Untersuchungen über die Vererbung derAlkaloidarmut und die Variabilität des Restalkaloidgehaltes beiLupinus albus L. Der Züchter 34:251–256. https://doi.org/10.1007/bf00705826

Przysiecka Ł, Książkiewicz M, Wolko B, Naganowska B (2015)Structure, expression profile and phylogenetic inference of chalconeisomerase-like genes from the narrow-leafed lupin (Lupinusangustifolius L.) genome. Front Plant Sci 6:268. https://doi.org/10.3389/fpls.2015.00268

Rogers DJ, Tanimoto TT (1960) A Computer program for classifyingplants. Science 132:1115–1118. https://doi.org/10.1126/science.132.3434.1115

Rychel S, KsiążkiewiczM, Tomaszewska M, Bielski W,Wolko B (2019)FLOWERING LOCUS T, GIGANTEA, SEPALLATA and FRIGIDAhomologs are candidate genes involved in white lupin (Lupinus

albus L.) early flowering. Mol Breed 39:43. https://doi.org/10.1007/s11032-019-0952-0

Šatović Z (1993) Biokemijske i genetske osobine alkaloida bijele lupine(Lupinus albus.). Sjemenarstvo 10:139–154

Shavrukov Y (2016) Comparison of SNP and CAPS markers applicationin genetic research in wheat and barley. BMC Plant Biol 16:11.https://doi.org/10.1186/s12870-015-0689-9

Sokal RR, Michener CD (1958) A statistical method for evaluating sys-tematic relationships. University of Kansas Scientific Bulletin 28:1409–1438

Stam P (1993) Construction of integrated genetic linkage maps by meansof a new computer package: Join Map. Plant J 3:739–744. https://doi.org/10.1111/j.1365-313X.1993.00739.x

Stracke S, Sato S, Sandal N, KoyamaM, Kaneko T, Tabata S, ParniskeM(2004) Exploitation of colinear relationships between the genomesof Lotus japonicus, Pisum sativum and Arabidopsis thaliana, forpositional cloning of a legume symbiosis gene. Theor Appl Genet108:442–449. https://doi.org/10.1007/s00122-003-1438-2

Święcicki W Developments in L. albus breeding. In: 4th InternationalLupin Conference, Geraldton, Western Australia, 1986.International Lupin association, Canterbury, New Zealand andDepartment of Agriculture, Western Australia, pp 14–19

Święcicki W, Jach K (1980) Variation and evolution of alkaloid complexin yellow lupine (Lupinus luteus L.) during domestication ActaAgrobotanica 33:177-195

Święcicki W, Święcicki WK (1995) Domestication and breeding im-provement of narrow-leafed lupin (L. angustifolius L.). J ApplGenet 36:155–167

Taylor CM et al (2019) INDEL variation in the regulatory region of themajor flowering time gene LanFTc1 is associated with vernalizationresponse and flowering time in narrow-leafed lupin (Lupinusangustifolius L.). Plant Cell Environ 42:174–187. https://doi.org/10.1111/pce.13320

Thiel T, Kota R, Grosse I, Stein N, Graner A (2004) SNP2CAPS: a SNPand INDEL analysis tool for CAPS marker development. NucleicAcids Res 32:e5–e5. https://doi.org/10.1093/nar/gnh006

Troll HJ (1958) Erbgänge des Alkaloidgehaltes und Beobachtungen überHeteros iswirkung bei Lupinus albus . Zei tschr i f t fürPflanzenzüchtung 39:35–46

Untergasser A, Nijveen H, Rao X, Bisseling T, Geurts R, Leunissen JAM(2007) Primer3Plus, an enhanced web interface to Primer3. NucleicAcids Res 35:W71–W74. https://doi.org/10.1093/nar/gkm306

Vipin CA et al (2013) Construction of integrated linkage map of a recom-binant inbred line population of white lupin (Lupinus albus L.).Breed Sci 63:292–300. https://doi.org/10.1270/jsbbs.63.292

von Sengbusch R (1942) Süßlupinen und Öllupinen. DieEntstehungsgeschichte einiger neuer Kulturpflanzen.Landwirtschaftliche Jahrbücher 91:719–880

Wang J et al (2017) Hierarchically aligning 10 legume genomes estab-lishes a family-level genomics platform. Plant Physiol 174:284–300. https://doi.org/10.1104/pp.16.01981

Wyrwa K, Książkiewicz M, Szczepaniak A, Susek K, Podkowiński J,Naganowska B (2016) Integration of Lupinus angustifolius L.(narrow-leafed lupin) genomemaps and comparative mappingwith-in legumes. Chromosome Res 24:355–378. https://doi.org/10.1007/s10577-016-9526-8

Yang H, Boersma JG, You M, Buirchell BJ, Sweetingham MW (2004)Development and implementation of a sequence-specific PCRmarker linked to a gene conferring resistance to anthracnose diseasein narrow-leafed lupin (Lupinus angustifolius L). Mol Breed 14:145–151. https://doi.org/10.1023/B:MOLB.0000038003.49638.97

Yang H, Renshaw D, Thomas G, Buirchell B, Sweetingham M(2008a) A strategy to develop molecular markers applicable toa wide range of crosses for marker assisted selection in plantbreeding: a case study on anthracnose disease resistance in lupin

280 (2019) 60:269–281J Appl Genetics

(Lupinus angustifolius L). Mol Breed 21:473–483. https://doi.org/10.1007/s11032-007-9146-2

Yang H, Shankar M, Buirchell J, Sweetingham W, Caminero C, Smith C(2002) Development of molecular markers using MFLP linked to agene conferring resistance to Diaporthe toxica in narrow-leafed lu-pin (Lupinus angustifolius L). Theor Appl Genet 105:265–270.https://doi.org/10.1007/s00122-002-0925-1

Yang H et al (2013) Draft genome sequence, and a sequence-definedgenetic linkage map of the legume crop species Lupinusangustifolius L. PLoS One 8:e64799. https://doi.org/10.1371/journal.pone.0064799

Yang S et al (2008b) Alfalfa benefits from Medicago truncatula: theRCT1 gene from M. truncatula confers broad-spectrum resistanceto anthracnose in alfalfa. Proc Natl Acad Sci U S A 105:12164–12169. https://doi.org/10.1073/pnas.0802518105

Yang Tet al (2017) Transcript profiling of a bitter variety of narrow-leafedlupin to discover alkaloid biosynthetic genes. J Exp Bot 68:5527–5537. https://doi.org/10.1093/jxb/erx362

You M, Boersma JG, Buirchell BJ, Sweetingham MW, Siddique KHM,Yang H (2005) A PCR-based molecular marker applicable formarker-assisted selection for anthracnose disease resistance in lupinbreeding. Cell Mol Biol Lett 10:123–134

Zhou G et al (2018) Construction of an ultra-high density consensusgenetic map, and enhancement of the physical map from genomesequencing in Lupinus angustifolius. Theor Appl Genet 131:209–223. https://doi.org/10.1007/s00122-017-2997-y

Publisher’s note Springer Nature remains neutral with regard tojurisdictional claims in published maps and institutional affiliations.

281(2019) 60:269–281Appl GeneticsJ